the treatment of the client by the worker is a key part of what is being bought and sold. To describe this aspect of the product, Hochschild coined the term emotional labor, which refers to the paid worker's managing her own feelings in order to create a certain state of mind in the customer.32
The second, though related, difference is in the central role consumers play in service labor processes. In manufacturing, goods are sold in a market distant from the factory, which might be in Detroit or in Bangladesh, so customers who buy these products never see the workers who make them. Not so with service clients, who are not only physically present as the interactive product is created but also, in fact, participate in its production. A dyadic relationship between managers and workers becomes a tripartite one among workers, managers, and clients.33 Interactive work also means that production and consumption occur simultaneously, linked in time just as they are brought together in space.
Self at Work
In regard to these characteristics—the importance of intangible workerclient interactions and the client's participation in the production of services—the literature on service work has tended to concentrate less on structure and class than on selfhood, interaction, and gender. First, scholars debate the effects that the performance of emotional labor has on workers who operate in what C. Wright Mills called “the personality market.”34 Hochschild drew on the Marxist notion of alienation to suggest that emotional laborers can be estranged from their emotions; subsequent work has looked at the pros and cons of emotional labor for workers in different settings.35 At a theoretical level, some analysts have discussed the broader social and cultural effects of commodifying and routinizing emotional labor, especially the possible loss of authenticity in human interaction generally.36 Despite research showing that emotional labor is not alienating for workers in all work situations, an implicit opposition between an “authentic,” agentic self and an estranged, alienated, and performative one often marks this literature, with authentic selfhood, for the most part, seen as located outside work settings.37 For example, researchers often talk of a self that needs to be “protected” at work or suggest that workers preserve their agentic selves through resistance.38 Many scholars have also looked at how service work occupations and interactions are gendered.39
A second focus of this research has been on how the self is managed in interactive work. Because the worker's presentation of self and capacity to interact constitute part of the product, managers must control the worker herself in order to control production. This requires extending managerial influence to areas that have traditionally been private, which can incur worker resistance. Depending on the setting, managers can try to routinize work by developing standard procedures, scripting interactions, and controlling customers, as they do at McDonald's.40 Or managers can attempt to transform workers’ selves in a more fundamental way, through extensive training and techniques, to ensure what Hochschild calls “deep acting.”41 (This approach is more appropriate in luxury production, in which workers have more autonomy, as we will see.) Furthermore, these scholars point out, service workers have another kind of manager: the client. Customer feedback and monitoring create a “second boss,” who supervises the worker even in the absence of managers.42
Service work researchers tend to cast both manager-worker and client-worker relations as antagonistic, involving subordination and struggles for control.43 As in traditional labor process theory, the language of “control” and “resistance” remains dominant. Workers are seen as responding to managerial dictates and customer demands either with passive compliance or active resistance, which ranges from refusing to smile to cursing at customers.44 Even when workers describe elements of their relations with customers as meaningful, scholars tend not to theorize the positive aspects of these contacts.45 Robin Leidner has offered a less pessimistic view, suggesting the possibility of shifting alliances among workers, managers, and customers, but few scholars since Leidner have looked closely at positive moments of client-worker relations. Furthermore, these studies rarely incorporate, methodologically or theoretically, the client's perspective.46
In the Marxist manufacturing paradigm, how managers and capitalists benefit from extracting workers’ labor power is clear: that extraction is the source of economic gain. Service work researchers have imported this sense of worker-manager antagonism into their view of the worker-client relationship, but they fail to theorize precisely how the client benefits from consuming workers’ labor and why this benefit is antithetical to workers’ interests. Partly this assumption has to do with the potential for individual customers to treat workers badly and with the belief that emotional labor has negative effects on workers, but these links are not often theoretically elaborated.47
In her ethnographic study of domestic servants and employers, Judith Rollins suggests that a moment of “psychological exploitation” stems from “the personal relationship between employer and employee.” She sees this exploitation as giving employers “the self-enhancing satisfactions that emanate from having the presence of an inferior.” Profit is not extracted, but the relation is exploitative because of its psychological benefit. Rollins also claims that domestic servants function to “validat[e] the employers’ lifestyle, ideology, and social world, from their familial interrelations to the economically and racially stratified system in which they live.”48 In this way, Rollins usefully theorizes interactive subordination and links it to structural inequality; however, she tends to map the interactive moment very closely onto the structural one, denying the possibility of positive interactions between workers and clients. Indeed, Rollins focuses primarily on workers’ deep resentment of and disdain for their client-employers, which she sees as stemming from this inequality. And although she implies that employers need the validation that workers provide, she does not investigate conflicts that employers might feel about their own entitlement to consume workers’ labor.
One could argue, alternatively, that structural inequality benefits workers and that workers are happy to serve clients more affluent than they. Some researchers argue that wealthy clients or employers treat workers better than less-rich ones.49 Another perspective suggests that workers identify with well-off clients and aspire to their lifestyle.50 These scholars tend not to use a class analysis, coding interactions as ways to mitigate “social distance” rather than unequal class relations.
Overall, despite significant progress in understanding interactive service work, most researchers have adopted the critical tone of labor process analysis without rethinking its theoretical basis in Marxist analysis of manufacturing production. This tendency, which also leads to dichotomous formulations of control and resistance, authentic and inauthentic selfhood, work and not-work selves, and production and consumption, has been limiting in several ways. First, the exploration of selfhood in interaction has been circumscribed by a focus on managerial dictates rather than on workplace relations and by the assumption that the authentic selfhood of workers arises outside work. Second, and related, consumers of services have primarily been seen as a source of constraint on workers rather than as a source of enjoyment or alliance or as a subject of study in their own right as participants in production. Finally, the literature has generally not theorized the classed nature of interactions or the links between structural inequality and interactive work. My study attempts to break down some of these limitations, looking at both the multiple ways workers and guests negotiate asymmetrical relations and the consequences of these negotiations for the reproduction of unequal entitlements to material resources and attention.
LOOKING AT LUXURY
Luxury service, in particular the luxury hotel, is a good place to explore in more depth the connections among work, class, and self. First, the worker's self is deeply implicated in the highly personalized and attentive service she provides. And it is a self-subordinating service, as guests’ every wish must be workers’ command. Research has established that workers value dignity on the job above most other considerations,51 but their dignity, in this case, seems constantly compromised by their subservience to guests. If any interactive workplace is likely to produce alienated and resentful workers, the luxury hotel is the one. On the other hand, the importance of service is a source of worker autonomy vis-à-vis managers, because the success of the guest's